Evaluation of Statistical Index Method in Flood Susceptibility Mapping

Flood is one of the natural catastrophes which has been occurring since ancient times. The main purpose of this study is assessing the statistical index methods used for flood mapping. Consequently, elevation, slope, land topography, the land wetness index, stream power, rainfall average, distance from the river, geology and landuse parameters of the Haraz watershed located in Mazandaran province were used. In order to prepare the landuse map of 2013, the images of Landsat Satellite, ENVI 5.1 and neural networks algorithm were utilized. The digital maps of all parameters were provided using ArcGIS software 10.1 and SAGA GIS 2 with Raster format. Then, the geographical positions of the 211 floodpoints of the region were prepared. These floodpoints were randomly divided into two groups, with 151 points (71%) and 61 points (30%) for calibration and validation, respectively. The set of calibration group points and effective parameters on the flood were introduced as the dependent and independent variables respevtively using the frequency ratio method. Then, the probability of the flood occurring for each class of parameters was calculated. At the end, the obtained weights for each class in the Geographical Information System (GIS) were applied to the corresponding layer and flood risk map of the studied region was prepared. The prediction accuracy of this method in order to prepare map within the region of Haraz is equal to 90 percent. The findings imply that the present technique for predicting the potential of flood risk is useful and reliable, especially for regions with no statistical data.